U.S. patent application number 11/063545 was filed with the patent office on 2005-06-30 for infrared ray detecting type imaging device.
Invention is credited to Funaki, Hideyuki, Iida, Yoshinori, Ikegawa, Sumio, Nakayama, Kohei, Shigenaka, Keitaro.
Application Number | 20050139774 11/063545 |
Document ID | / |
Family ID | 29230281 |
Filed Date | 2005-06-30 |
United States Patent
Application |
20050139774 |
Kind Code |
A1 |
Ikegawa, Sumio ; et
al. |
June 30, 2005 |
Infrared ray detecting type imaging device
Abstract
An imaging device comprises a select line, a first signal line
crossing the select line, and a first pixel provided at a portion
corresponding to a crossing portion of the select line and the
first signal line, the first pixel comprising a first buffer layer
formed on a substrate, a first bolometer film formed on the first
buffer layer, made of a compound which undergoes metal-insulator
transition, and generating a first temperature detection signal, a
first switching element formed on the substrate, selected by a
select signal from the select line, and supplying the first
temperature detection signal to the first signal line, and a metal
wiring connecting a top surface of the first bolometer film to the
first switching element.
Inventors: |
Ikegawa, Sumio;
(Musashino-shi, JP) ; Nakayama, Kohei;
(Kawasaki-shi, JP) ; Funaki, Hideyuki; (Tokyo,
JP) ; Iida, Yoshinori; (Tokyo, JP) ;
Shigenaka, Keitaro; (Hachioji-shi, JP) |
Correspondence
Address: |
OBLON, SPIVAK, MCCLELLAND, MAIER & NEUSTADT, P.C.
1940 DUKE STREET
ALEXANDRIA
VA
22314
US
|
Family ID: |
29230281 |
Appl. No.: |
11/063545 |
Filed: |
February 24, 2005 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11063545 |
Feb 24, 2005 |
|
|
|
10392826 |
Mar 21, 2003 |
|
|
|
Current U.S.
Class: |
250/338.1 ;
257/E27.008; 348/E5.09 |
Current CPC
Class: |
H01L 27/16 20130101;
G01J 5/24 20130101; H04N 5/33 20130101; G01J 2005/202 20130101;
G01J 5/20 20130101; G01J 5/22 20130101 |
Class at
Publication: |
250/338.1 |
International
Class: |
G01J 005/00 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 22, 2002 |
JP |
2002-081795 |
Claims
1-10. (canceled)
11. An imaging device comprising: a first select line; a first
signal line crossing the first select line; a first pixel provided
at a portion corresponding to a crossing portion of the first
select line and the first signal line, the first pixel comprising a
first bolometer film generating a first temperature detection
signal, and a first switching element selected by a first select
signal from the first select line and supplying the first
temperature detection signal to the first signal line; a second
signal line crossing the first select line; a second pixel provided
at a portion corresponding to a crossing portion of the first
select line and the second signal line, the second pixel comprising
a second bolometer film generating a second temperature detection
signal, and a second switching element selected by the first select
signal and supplying the second temperature detection signal to the
second signal line; and a control circuit controlling a width of
the first select signal in accordance with the second temperature
detection signal.
12. The imaging device according to claim 11, wherein the first
bolometer film does not detect infrared rays.
13. The imaging device according to claim 11, wherein the control
circuit includes a comparator comparing a voltage of the second
temperature detection signal with a predetermined voltage to
generate the first select signal.
14. The imaging device according to claim 11, further comprising: a
second select line crossing the first signal line and the second
signal line; a third pixel provided at a portion corresponding to a
crossing portion of the second select line and the first signal
line, the third pixel comprising a third bolometer film generating
a third temperature detection signal, and a third switching element
selected by a second select signal from the second select line and
supplying the third temperature detection signal to the first
signal line; and a fourth pixel provided at a portion corresponding
to a crossing portion of the second select line and the second
signal line, the fourth pixel comprising a fourth bolometer film
generating a fourth temperature detection signal, and a fourth
switching element selected by the second select signal and
supplying the fourth temperature detection signal to the second
signal line; wherein the control circuit is configured to control a
width of the second select signal in accordance with the fourth
temperature detection signal.
15. The imaging device according to claim 14, wherein the fourth
bolometer film does not detect infrared rays.
16. The imaging device according to claim 11, wherein a plurality
of first select lines cross a plurality of first signal lines and
the second signal line, and a plurality of first pixels are
provided at portions corresponding to crossing portions of the
first select lines and the first signal lines, and a plurality of
the second pixels are provided at portions corresponding to
crossing portions of the first select lines and the second signal
line.
17-22. (canceled)
23. The imaging device according to claim 11, wherein each of the
first and second bolometer films is made of a compound which
undergoes metal-insulator transition.
24. The imaging device according to claim 23, wherein the compound
is expressed by RNiO.sub.3-d, where R is at least one element
selected from Pr, Nd, Sm, Eu and Bi, and d is a value showing
deviation from stoichiometry.
25. The imaging device according to claim 24, wherein the value of
d ranges from -0.1, inclusive, to 0.2, inclusive.
26. The imaging device according to claim 23, wherein the compound
is expressed by Ca.sub.2-xSr.sub.xRuO.sub.4-d, where d is a value
showing deviation from stoichiometry, and x ranges from 0,
inclusive, to 0.05, inclusive, or by Ca.sub.2-x'RuO.sub.4-d', where
d' is a value showing deviation from stoichiometry, and x' is
greater than 0 and smaller than 0.32.
27. The imaging device according to claim 26, wherein the value of
d ranges from -0.1, inclusive, to 0.2, inclusive, and the value of
d' ranges from -0.1, inclusive, to 0.2, inclusive.
28. The imaging device according to claim 11, wherein a maximum
absolute value of a temperature coefficient of resistance of each
of the first and second bolometer films is more than 3%.
29. The imaging device according to claim 11, wherein each of the
first and second bolometer films is made of an epitaxial film.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is based upon and claims the benefit of
priority from the prior Japanese Patent Application No. 2002-81795,
filed Mar. 22, 2002, the entire contents of which are incorporated
herein by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention relates to an imaging device and a
method of manufacturing the same, and more particularly to an
infrared ray detecting type imaging device (an infrared imaging
device) and a method of manufacturing the same.
[0004] 2. Description of the Related Art
[0005] An infrared imaging device is proposed, in which a bolometer
film is formed on a semiconductor substrate on which a circuit
including transistor and wiring and the like is formed. This device
detects an infrared ray for each pixel by making use of change of a
resistance value of the bolometer film depending on temperature,
and reads the detected signal by way of the transistor. In the
conventional infrared imaging device, first, a circuit portion
including a transistor and a metal wiring such as an aluminum
wiring is formed on a semiconductor substrate, and then a bolometer
film is formed thereon.
[0006] In such an infrared imaging device, however, the process
temperature when forming a bolometer film must be about 450.degree.
C. or less, preferably about 400.degree. C. or less. If the process
temperature is higher, the metal wiring such as an aluminum wiring
deteriorates. Further, by heat treatment at high temperature of
about 800.degree. C. or more, the transistor characteristic also
deteriorates. Conventional films such as vanadium oxide films can
be formed at relatively low temperature, and there is no problem,
but when a material requiring to be formed at high temperature is
used as a bolometer film, the metal wiring and transistors
deteriorate. Therefore, in the conventional infrared imaging
device, materials usable for the bolometer film are limited.
[0007] Hitherto, moreover, there are undulations due to wiring
steps and contact holes beneath the bolometer film since the metal
wiring and the like are formed on a lower side of the bolometer
film. At these positions of steps and contact holes, the bolometer
film has crystal disturbance and grain boundary, which causes noise
or characteristic deterioration.
[0008] Therefore, in the conventional infrared imaging device,
since the bolometer film must be formed at a relatively low
temperature, materials for the bolometer film are limited. Besides,
by the undulations existing beneath the bolometer film, noise and
characteristic deterioration are caused. It was hence difficult to
obtain an infrared imaging device of high performance.
BRIEF SUMMARY OF THE INVENTION
[0009] A first aspect of the invention, there is provided an
imaging device comprising: a select line; a first signal line
crossing the select line; and a first pixel provided at a portion
corresponding to a crossing portion of the select line and the
first signal line; the first pixel comprising: a first buffer layer
formed on a substrate; a first bolometer film formed on the first
buffer layer, made of a compound which undergoes metal-insulator
transition, and generating a first temperature detection signal; a
first switching element formed on the substrate, selected by a
select signal from the select line, and supplying the first
temperature detection signal to the first signal line; and a metal
wiring connecting a top surface of the first bolometer film to the
first switching element.
[0010] A second aspect of the invention, there is provided an
imaging device comprising: a first select line; a first signal line
crossing the first select line; a first pixel provided at a portion
corresponding to a crossing portion of the first select line and
the first signal line, the first pixel comprising a first bolometer
film generating a first temperature detection signal, and a first
switching element selected by a first select signal from the first
select line and supplying the first temperature detection signal to
the first signal line; a second signal line crossing the first
select line; a second pixel provided at a portion corresponding to
a crossing portion of the first select line and the second signal
line, the second pixel comprising a second bolometer film
generating a second temperature detection signal, and a second
switching element selected by the first select signal and supplying
the second temperature detection signal to the second signal line;
and a control circuit controlling a width of the first select
signal in accordance with the second temperature detection
signal.
[0011] A third aspect of the invention, there is provided a method
of manufacturing an imaging device, comprising: forming a buffer
layer on a substrate; forming a bolometer film made of a compound
which undergoes metal-insulator transition, on the buffer layer;
forming a switching element on the substrate after forming the
bolometer film; and forming a metal wiring to connect the bolometer
film to the switching element after forming the bolometer film.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING
[0012] FIG. 1 is a diagram showing an equivalent circuit of an
infrared imaging device according to an embodiment of the
invention.
[0013] FIG. 2 is a view showing a sectional structure of a pixel of
the infrared imaging device according to the embodiment of the
invention.
[0014] FIG. 3 is a view showing a plane structure of the pixel of
the infrared imaging device according to the embodiment of the
invention.
[0015] FIG. 4A to FIG. 4C are views showing a method of
manufacturing an infrared imaging device according to an embodiment
of the invention.
[0016] FIG. 5 is a view showing a sectional structure of a pixel in
a modified example of the infrared imaging device according to the
embodiment of the invention.
[0017] FIG. 6 is a view showing a sectional structure of a pixel in
another modified example of the infrared imaging device according
to the embodiment of the invention.
[0018] FIG. 7 is a diagram showing a circuit structure of an entire
infrared imaging device according to the embodiment of the
invention.
[0019] FIG. 8 is a view showing signal waveforms of the circuit
shown in FIG. 7.
[0020] FIG. 9 is a view showing the driving principle of the
infrared imaging device according to the embodiment of the
invention.
[0021] FIG. 10 is a schematic diagram of a molecular beam epitaxy
apparatus for use in manufacturing the infrared imaging device
according to the embodiment of the invention.
[0022] FIG. 11 is a view showing temperature dependence of
resistivity and temperature dependence of TCR, in the bolometer
film of the infrared imaging device according to the embodiment of
the invention.
[0023] FIG. 12 is a view showing substrate temperature dependence
of TCR when using a LaAlO.sub.3 substrate, in the bolometer film of
the infrared imaging device according to the embodiment of the
invention.
[0024] FIG. 13 is a view showing O.sub.3/Ni supply ratio dependence
of TCR, in the bolometer film of the infrared imaging device
according to the embodiment of the invention.
[0025] FIG. 14 is a view showing Ni/Sm composition ratio dependence
of TCR, in the bolometer film of the infrared imaging device
according to the embodiment of the invention.
[0026] FIG. 15 is a view showing annealing temperature dependence
of XRD intensity, in the bolometer film of the infrared imaging
device according to the embodiment of the invention.
[0027] FIG. 16 is a diagram showing oxygen partial pressure
dependence of XRD intensity, in the bolometer film of the infrared
imaging device according to the embodiment of the invention.
[0028] FIG. 17 is a view showing temperature dependence of
resistivity and temperature dependence of TCR, in the bolometer
film of the infrared imaging device according to the embodiment of
the invention.
DETAILED DESCRIPTION OF THE INVENTION
[0029] First of all, general matters about the embodiment of the
invention will be described below.
[0030] The performance of a bolometer film is generally expressed
by TCR (temperature coefficient of resistance). Assuming the
resistance of the bolometer film at temperature T to be R, TCR is
expressed as follows.
TCR=(1/R)(dR/dT)
[0031] For infrared detection at higher sensitivity than before, it
should be preferably .vertline.TCR.vertline.>3%/K, and more
preferably .vertline.TCR.vertline.>4%/K. For lower cost and
higher resolution, the pixel pitch should be smaller than before,
for example, about 15 .mu.m. However, when the pixel pitch is
reduced, the incident thermal energy in one pixel is reduced. As a
result, the sensitivity is lowered, and the value of NETD (noise
equivalent temperature difference) increases. When using the
bolometer film in a high sensitivity infrared camera, the NETD
value is preferred to be 60 to 100 mK. To achieve the NETD value of
60 to 100 mK at the pixel pitch of 15 .mu.m, it is difficult with
the conventional vanadium oxide bolometer film. That is, to achieve
the NETD value of 60 to 100 mK at the pixel pitch of 15 .mu.m, the
bolometer sensitivity must be not less than two times higher than
conventional one. Besides, since the bolometer temperature is
raised more than the room temperature by the pulse bias current
when measuring the bolometer resistance, it is preferred to realize
.vertline.TCR.vertline.&- gt;3%/K in a temperature range of 300
to 350 K.
[0032] Bolometer materials having such characteristic include
compound crystals showing metal-insulator transition, in
particular, the following two types of compound crystals.
[0033] (1) RNiO.sub.3-d (where R is at least one element selected
from Pr, Nd, Sm, Eu and Bi, and d is a value showing deviation from
the stoichiometry of oxygen, which is usually
-0.1.ltoreq.d.ltoreq.0.2). A representative example is
Sm.sub.1-xA.sub.yNi.sub.yO.sub.3-d (where A is Nd or Bi,
0.ltoreq.x.ltoreq.0.5, 0.9<y<1.1).
[0034] (2) Ca.sub.2-xSr.sub.xRuO.sub.4-d (where d is a value
showing deviation from the stoichiometry of oxygen, which is
usually -0.1.ltoreq.d.ltoreq.0.2, 0.ltoreq.x.ltoreq.0.05), or
Ca.sub.2-xRuO.sub.4-d (where d is a value showing deviation from
the stoichiometry of oxygen, which is usually
-0.1.ltoreq.d.ltoreq.0.2, 0<x<0.32).
[0035] In these bolometer materials, the TCR value at around
metal-insulator transition is sufficiently large so as to obtain an
infrared imaging device of high sensitivity. In these bolometer
materials, by optimizing the forming condition and composition, the
metal-insulator transition occurs at a temperature suited to a
non-cooled type infrared imaging device (T.sub.MI=320 to 410 K).
Herein, the T.sub.MI is the metal-insulator transition temperature,
and it is defined by the temperature at which the sign of the TCR
is changed.
[0036] In bulk SmNiO.sub.3, metal-insulator transition occurs at
T.sub.MI=403 K, and the T.sub.MI is lowered when part of Sm is
replaced by Nd, as disclosed by J. B. Torrance et al. in Phys. Rev.
B, 45, p. 8209 (1992). The present inventor has succeeded in
achieving metal-insulator transition at the room temperature or
more for the first time in a thin film of RNiO.sub.3-d, and has
made it possible to apply to the infrared imaging device.
[0037] As a result of experiment by using
Sm.sub.1-xA.sub.xNi.sub.yO.sub.3- -d having the perovskite
structure, in order to obtain .vertline.TCR.vertline.>3%/K at
the room temperature or more, it was found that the following
condition is needed. Ultimately, a .vertline.TCR.vertline. value of
more than 6%/K was obtained at the room temperature or more.
[0038] (1) Concerning deviation of composition ratio of site A
element and site B element in the perovskite structure, 0.9<y is
needed.
[0039] (2) The required film forming temperature is 550.degree. C.
or more.
[0040] (3) In the molecular beam epitaxy method, when O.sub.3 gas
is used as oxidizing gas, the O.sub.3 flux is required to be not
less than 30 times of Ni flux.
[0041] (4) When the underlying layer is SrTiO.sub.3 or NdGaO.sub.3,
the metal-insulator transition rarely occurs, and when the
underlying layer is LaAlO.sub.3, the metal-insulator transition is
successfully obtained.
[0042] (5) The SmNi.sub.yO.sub.3-d film has T.sub.MI of 400 to 410
K, and the transition temperature is relatively high. By lowering
the T.sub.MI, holding temperature of the element can be brought
closer to the room temperature, and it is easier to use and the
cost is lowered. To lower the T.sub.MI, it is found effective to
replace part of Sm with Bi (A=Bi). Substitution amount x is
preferred to be 0<x<0.09. Since Bi and Bi oxide are low in
melting point, the process temperature can be lowered by replacing
part of Sm with Bi.
[0043] In bulk Ca.sub.2RuO.sub.4, metal-insulator transition occurs
at T.sub.MI=357 K, as disclosed by C. S. Alexander et al. in Phys.
Rev. B, 60, p. 8422 (1999). By replacing part of Ca with La or Sr,
the T.sub.MI and resistivity are lowered, as disclosed by G. Gao et
al. in Phys. Rev. B, 61, p. 5053 (2000). The present inventor has
succeeded in achieving of metal-insulator transition for the first
time in a thin film of Ca.sub.2RuO.sub.4, and has made it possible
to apply to the infrared imaging device.
[0044] As a result of experiment by using Ca.sub.2-xRuO.sub.4-d
having the layered perovskite structure, in order to obtain the
metal-insulator transition, it was found that the following
condition is needed.
[0045] (1) When the underlying layer is SrTiO.sub.3 or NdGaO.sub.3,
the metal-insulator transition rarely occurs, and when the
underlying layer is LaAlO.sub.3, the metal-insulator transition is
successfully obtained.
[0046] (2) First an amorphous film is formed, and then a heat
treatment is performed to obtain a desired crystal structure.
[0047] (3) To obtain a desired crystal structure, it is required to
heat in a mixed atmosphere of inert gas and oxygen gas of 0.05% or
more and less than 1%, in a temperature range of 990.degree. C. and
1050.degree. C.
[0048] (4) Before this heat treatment, it is preferred to heat for
10 hours or more in oxygen gas atmosphere at 700 to 800.degree.
C.
[0049] (5) Ca.sub.2RuO.sub.4 has a relatively high phase transition
temperature of T.sub.MI=357 K. By lowering the T.sub.MI, holding
temperature of the element can be brought closer to room
temperature, and it is easier to use and the cost is lowered. To
lower the T.sub.MI, it is effective to lose part of Ca. The loss
amount x is preferred to be 0<x<0.32. By lowering the
T.sub.MI by Ca loss, the T.sub.MI can be adjusted without
introducing Sr.
[0050] In these two types of bolometer materials, in order to
obtain a desired crystal structure for achieving metal-insulator
transition, a high temperature process at 450.degree. C. or more is
needed. Accordingly, the conventional method of manufacturing an
infrared imaging device could not be applied. In this embodiment,
before forming ROIC (read-out integrated circuit) including
transistor and metal wiring on a semiconductor substrate, a
bolometer film is formed. As a result, the high performance of
these materials can be utilized.
[0051] Further, in these two materials, in order to achieve
metal-insulator transition, selection of the underlying layer is
important. It is preferred that first, a buffer layer is formed on
the semiconductor substrate (Si substrate), and then a bolometer
film is formed thereon. It is also preferred to form a buffer layer
in two layers as described below.
[0052] A first layer is preferably a thin film of oxide epitaxially
grown on a silicon substrate. The crystal structure of this oxide
is preferably the perovskite structure, fluorite structure, or
C-type rare earth structure. It is also preferred that the lattice
of a first buffer layer is matched to a certain extent with the
lattice of a second buffer layer, and the lattice mismatch is
preferably within .+-.10%. For example, the first buffer layer is
SrTiO.sub.3 (100) orientation film, CeO.sub.2 (100) orientation
film, or RE.sub.2O.sub.3 (100) orientation film (where RE is a
trivalent rare-earth element or Y), epitaxially grown on the Si
(100) substrate.
[0053] A second buffer layer is preferred to be a thin film of
oxide epitaxially grown on the first buffer layer. The crystal
structure of this oxide is preferred to belong to perovskite
family. The lattice of the second buffer layer is preferred to be
matched sufficiently with the lattice of a bolometer film, and the
lattice mismatch is preferably within .+-.2.5%. For example, the
second buffer layer is a LaAlO.sub.3 film. The thickness of the
second buffer layer is preferred to be thick enough to obtain its
proper lattice constant without having effect of the lattice
constant of the first buffer layer, and is preferably 50 nm or more
(more preferably 100 nm or more).
[0054] In this embodiment, as the buffer layer, first, the
SrTiO.sub.3 (100) orientation film is epitaxially grown on the Si
(100) substrate, and then the LaAlO.sub.3 (100) orientation film is
grown epitaxially. The technology of direct epitaxial growth of
SrTiO.sub.3 on the Si (100) substrate is disclosed, for example, by
R. A, McKee et al., Phys. Rev. Lett. 81, p. 3014 (1998). The
SrTiO.sub.3 (100) orientation film functions as a seed layer for
epitaxial growth of perovskite oxide on Si. Accordingly, the
thickness of the SrTiO.sub.3 (100) orientation film is enough at 3
unit cells or more, and typically it is 2 nm. The LaAlO.sub.3 (100)
orientation film is required to have a surface lattice constant
closer to the lattice constant of bulk such that the lattice
mismatch to the bolometer film is a proper value. From such
viewpoint, the thickness of the LaAlO.sub.3 (100) orientation film
is preferred to be 50 nm or more, and typically it is 100 nm.
[0055] When reading a signal from the infrared imaging device, the
resistance of the bolometer film is measured by passing pulse
current in the bolometer film. Assuming the hold temperature of the
bolometer film to be T.sub.S, the self-heating temperature dT.sub.S
by reading current is 3 to 70 K (typically 10 to 20 K). By
contrast, the temperature rise dT.sub.IR by infrared ray is the
order of mK. In metal-insulator transition, as shown in FIG. 11,
the TCR has a temperature dependence, and the temperature rang
allowing large absolute values of TCR is narrow. Therefore, to
measure at high sensitivity, the temperature T.sub.P where the
absolute value of TCR reaches the peak should be somewhere between
T.sub.S and T.sub.S+dT.sub.S. The peak temperature T.sub.P is
preferred to be about the middle point of T.sub.S and
T.sub.S+dT.sub.S, or slightly closer to T.sub.S+dT.sub.S rather
than T.sub.S. To satisfy such condition, the pulse width of reading
current is preferred to be adjusted in every device or in every
line within the device. The device is preferred to incorporate a
pixel not sensitive to infrared ray (insensitive pixel). In this
case, the resistance change of the bolometer film in the
insensitive pixel is detected, and the reading pulse width in
ordinary pixel is determined on the basis of the detection signal.
As a result, it is possible to detect always near the peak
temperature T.sub.P, and an imaging device of high sensitivity is
realized.
[0056] To detect at high sensitivity in an infrared imaging device,
the pulse current is preferred to be 10 to 100 .mu.A, and the
voltage generated by the detection pulse is preferred to be 1 to
10V. Therefore, the resistance value of the bolometer film in one
pixel is preferably 10 to 100 k.OMEGA.. As shown in FIG. 11, the
resistivity of Sm.sub.1-xA.sub.xNi.sub.YO.sub.3-d is about
5.times.10.sup.-4 to 5.times.10.sup.-3 .OMEGA.cm. A proper
thickness of the bolometer film from the view points of sharp
metal-insulator transition and proper thermal conductance is in a
range of 30 to 200 nm, and it is typically 50 nm. Therefore, in
order that the length/width ratio of the bolometer film may be 10
to 1000, as shown in FIG. 3 later, the bolometer film is preferred
to be processed in a meandering shape. In the example shown later,
the length/width ratio is about 47. In the example of
Ca.sub.2-xSr.sub.xRuO.sub.4-d shown in FIG. 17, the resistivity is
about 4.times.10.sup.-3 to 1.times.10.sup.-2 .OMEGA.cm. A proper
thickness of the bolometer film from the view points of sharp
metal-insulator transition and proper thermal conductance is in a
range of 30 to 200 nm, and it is typically 80 nm. It is hence
desired to process the bolometer film in a meandering shape such
that the length/width ratio of the bolometer is 8 to 200.
[0057] Referring now to the drawings, embodiments of the invention
are described in detail below.
[0058] FIG. 1 is an equivalent circuit diagram of a non-cooled type
infrared imaging device of the embodiment. Each pixel portion has a
temperature detecting portion (heat sensitive portion) 41 using a
bolometer film, and a MIS transistor (switching element) 42. A
select line 43 is connected to the gate of each transistor 42
provided in the row direction, and a read line (signal line) 44 is
connected to the drain of each transistor 42 provided in the column
direction.
[0059] FIG. 2 is a sectional view of the pixel portion of the
infrared imaging device.
[0060] In the example shown in FIG. 2, an insulating layer 12 is
formed on a silicon substrate (semiconductor substrate) 11 which is
a support substrate, and a silicon layer (semiconductor layer) 13
is formed on the insulating layer 12, thereby composing a so-called
SOI substrate. Further, a buffer layer 14 of an insulating layer is
formed on the silicon layer 13, and a bolometer film 15 is formed
on the buffer layer 14. The bolometer film 15 is formed by using
any of the materials mentioned above, that is,
Sm.sub.1-xA.sub.xNi.sub.yO.sub.3-d, Ca.sub.2-xSr.sub.xRuO.sub.4-d,
or Ca.sub.2-xRuO.sub.4-d. The buffer layer 14 is a stacked film of
SrTiO.sub.3 and LaAlO.sub.3.
[0061] Beneath the bolometer film 15, a hollow space 16 is formed
by removing part of the silicon substrate 11. This hollow space 16
is for thermally isolating the bolometer film 15. A MIS transistor
portion (MIS transistor forming region) 17 is provided on the
silicon layer 13. A metal wiring (for example, Al wiring) 18 is
connected to one end of the bolometer film 15, and by this metal
wiring 18, the bolometer film 15 and the source of the MIS
transistor 17 are connected with each other. The other end of the
bolometer film 15 is grounded by way of the metal wiring 18.
[0062] As shown in FIG. 2, the metal wiring 18 contacts with the
top of the bolometer film 15. That is, the bolometer 15 has been
already formed before forming the metal wiring. Accordingly, unlike
the prior art, the bolometer film is not formed on the steps or
undulations of the metal wiring, but is formed on a flat surface of
the buffer layer 14. It is hence free from disorder of crystal or
grain boundary in the bolometer film due to steps or undulations.
Therefore, the bolometer film excellent in crystallinity is
obtained, and occurrence of noise and deterioration of
characteristic can be prevented.
[0063] FIG. 3 shows a plane structure of the pixel portion of the
infrared imaging device having the basic structure as shown in FIG.
2.
[0064] In FIG. 3, reference numeral 21 is a SOI substrate, 22 is a
hollow pattern, 23 is a meandering bolometer film pattern, 24 is a
MIS transistor portion (MIS transistor forming region), and 25 to
28 are wirings. The wiring 26 is for connecting between one end of
the bolometer film 23 and source of the MIS transistor 24. The
wiring 25 is for grounding the other end of the bolometer film 23.
The wiring 27 corresponds to a select line, and is connected to the
gate of the MIS transistor 24. The wiring 28 corresponds to a read
line (signal line), and is connected to the drain of the MIS
transistor 24.
[0065] The size of one pixel is, for example, about 50
.mu.m.times.50 .mu.m to 15 .mu.m.times.15 .mu.m. A smaller chip
area leads to reduction of cost, and there is an increasing demand
for high resolution and multiple pixels, thereby the size of one
pixel is preferred to be about 15 .mu.m.times.15 .mu.m. Since the
wavelength of the infrared ray to be detected is about 8 to 14
.mu.m, it is meaningless to define the pixel pitch of 10 .mu.m or
less from the viewpoint of diffraction limit. The number of pixels
is, for example, 320.times.240, and 640.times.480, for example,
where a high resolution is demanded.
[0066] As shown in FIG. 3, when the bolometer film portion
(detecting portion) and transistor portion are formed adjacently to
each other, the rate of the detecting portion area to the pixel
area (fill factor) becomes smaller. In the example in FIG. 3, the
fill factor is about 25%. When the fill factor drops, the
sensitivity decreases. To compensate for this loss, it is effective
to form an infrared ray absorption portion of umbrella structure
above the substrate in every pixel. This is disclosed in Japanese
Patent No. 3040356. By employing this technology, the fill factor
can be improved to 90% or more, nearly to 100%, and an infrared
imaging device of high sensitivity is obtained.
[0067] FIG. 4A to FIG. 4C show a method of manufacturing the
infrared imaging device shown in FIG. 2.
[0068] First, as shown in FIG. 4A, on the SOI substrate composed of
the silicon substrate 11, insulating film 12 and silicon layer 13,
the buffer layer 14 (stacked film of SrTiO.sub.3/LaAlO.sub.3) is
grown epitaxially. In succession, on the buffer layer 14, the
bolometer film 15 (using Sm.sub.1-xA.sub.xNi.sub.yO.sub.3-d,
Ca.sub.2-xSr.sub.xRuO.sub.4-d, or Ca.sub.2-xRuO.sub.4-d) is formed
in a high temperature process as mentioned above.
[0069] Next, as shown in FIG. 4B, the bolometer film 15 and buffer
layer 14 are processed in a required shape. Further, the transistor
17 is formed on the silicon layer 13, and further the metal wiring
(for example, Al wiring) 18 is formed. By this metal wiring 18, the
bolometer film 15 and transistor 17 are connected with each
other.
[0070] As shown in FIG. 4C, part of the silicon substrate 11 is
removed by anisotropic etching, the hollow space 16 having such a
pattern that includes the pattern of the bolometer film 15. At this
time, the insulating layer 12 functions as an etching stopper.
[0071] In this embodiment, the bolometer film 15 is formed in the
high temperature process as mentioned above, and this bolometer
film 15 is formed before formation of the transistor 17 and metal
wiring 18. Accordingly, the transistor 17 and metal wiring 18 are
not exposed to high temperature in the process of forming the
bolometer film 15. Therefore, when the material requiring high film
forming temperature is used in the bolometer film, unlike the prior
art, it is free from deterioration of metal wiring or transistor
characteristic, so that an infrared imaging device excellent in
performance is obtained.
[0072] FIG. 5 shows another example of the infrared imaging device
in the embodiment of the invention. Basically it is same as
explained above, except that a bulk Si substrate 10 is used instead
of the SOI substrate. In this example, the buffer layer 14 is used
as an etching stopper when forming the hollow space 16 by
anisotropic etching. The anisotropic etching is performed by a wet
etching process using tetramethyl ammonium hydroxide or the like as
an etchant. When the buffer layer 14 is used as an etching stopper,
a thick buffer layer is needed in order to support the hollow
structure. Hence, the thickness of the buffer layer 14 is about 0.5
.mu.m or more, preferably about 0.8 .mu.m.
[0073] Thus, in the example shown in FIG. 5, since the buffer layer
14 is used as an etching stopper, an ordinary inexpensive bulk Si
substrate can be used as compared with the SOI substrate, and the
manufacturing cost can be reduced.
[0074] FIG. 6 shows another example of the infrared imaging device
in the embodiment of the invention. Basically it is same as
explained above, except that a bulk Si substrate 10 is also used
instead of the SOI substrate. Further in this example, an etching
stop layer 19 formed of a silicon oxide film (SiO.sub.2 film) or
the like is provided. By using this etching stop layer 19 and
buffer layer 14 as etching stoppers, the hollow space 16 is formed
by isotropic etching. The isotropic etching is performed by a dry
etching process using XeFe.sub.2 or the like as etching gas.
[0075] It is also possible to use a SON (silicon on nothing)
substrate. The method of fabricating the SON substrate is disclosed
by Ichiro Mizushima et al. in Applied Physics, October 2000, p.
1187 (in Japanese, published by Japanese society of applied
physics). By forming a trench in a bulk Si substrate and heating in
hydrogen atmosphere at about 1100.degree. C., an Empty Space in
Silicon (ESS) can be formed. By applying this technique, a hollow
structure (hollow space) can be formed.
[0076] An example of a method of driving the infrared imaging
device according to the embodiment of the invention will be
explained by referring to FIG. 7 to FIG. 9. FIG. 7 is a diagram
showing an example of the infrared imaging device including
peripheral circuits such as driving circuit, FIG. 8 is a view
showing signal waveforms of the circuits shown in FIG. 7, and FIG.
9 is a view showing the driving principle of the infrared imaging
device. In FIG. 8, the axis of abscissas is the time and the axis
of ordinates represents the voltage.
[0077] The resistance of the bolometer film varies with temperature
changes due to infrared irradiation. Optimization of reading pulse
width, when reading a signal corresponding to such resistance
changes, is explained by referring to FIG. 9. The pulse width is
typically about 10 to 100 .mu.sec.
[0078] Schematically, (a) in FIG. 9 shows the reading current
flowing in the bolometer film, (b) in FIG. 9 shows voltage changes
of the bolometer film caused by the reading current, and (c) in
FIG. 9 shows temperature changes of the bolometer film caused by
the reading current.
[0079] When a current flows in the bolometer, the temperature (c)
of the bolometer film gradually elevates by self-heating. Using a
bolometer film of TCR<0, the voltage (b) applied between both
ends of the bolometer film decreases with increasing temperature.
The dotted line in (c) schematically shows temperature changes of
the bolometer film in the case of a continuous incidence of
infrared ray into the pixel portion. The temperature rise based on
infrared ray is dT.sub.IR. Assuming the initial temperature (hold
temperature) of the bolometer film to be T.sub.S, the temperature
rise caused by self heating by reading current pulse is dT.sub.S.
In metal-insulator transition, for example as explained later in
FIG. 11, the TCR is dependent on temperature. It is hence desired
to optimize the reading current pulse width such that the
temperature T.sub.P where the absolute value of the TCR reaches the
peak is the optimum temperature between T.sub.S and
T.sub.S+dT.sub.S. The hold temperature T.sub.S varies with the
reading current value and pulse width, thermal time constant of
heat sensitive part (detecting portion), and ambient temperature.
The peak temperature T.sub.P may fluctuate between devices or in a
device. It is hence desired to adjust the reading current pulse
width for each device or for each row line in the device. A
specific method will be explained below.
[0080] In FIG. 7, the basic structure of the pixel portion
including the detecting portion 41 using the bolometer film and MIS
transistor 42, and the basic structure of the select line 43 and
read line (signal line) 44 are as already explained. In this
example, plural pixel portions provided in predetermined column are
insensitive pixel column line (insensitive pixel group) 45. One
method to make insensitive pixels is overcoating a metal reflection
plate on the pixels to avoid incidence of infrared rays into the
detecting portions 41.
[0081] The transistors 42 are selected sequentially by AND gates
53, each outputs operation result c of output b of a row select
circuit 51 and output a of a comparator 52. The output of the
comparator 52 is connected also to the AND gates 54. Current
sources 55 are connected to the input portions of read lines
(signal lines) 44, and transistors 56 are connected to the output
portions of the read lines. The outputs of the transistors 56 are
connected to transistors 57, capacitors 58, and transistors 59. The
transistors 59 are sequentially selected by a control signal from a
column select circuit 60.
[0082] Referring now to the timing chart shown in FIG. 8, the
circuit operation in FIG. 7 is explained.
[0083] First, a reset signal V.sub.res is applied to each
transistor 57. As a result, each capacitor 58 is charged with a
power supply voltage V.sub.d through each transistor 57 being
turned on. Terminal voltage V.sub.ob of the capacitor 58
corresponding to the insensitive pixel column line is supplied to
the positive terminal of the comparator 52, and hence the output a
of the comparator 52 becomes high level. A reference voltage
V.sub.c is supplied to the negative terminal of the comparator 52.
This reference voltage V.sub.c is predetermined for each device so
that the pulse width of the read pulse c being output from the AND
gate 53 may be optimized.
[0084] In a specific time after supply of reset signal V.sub.res,
select signal b is supplied to the AND gate 53 from the row select
circuit 51. A select signal V.sub.g is supplied to the AND gate 54.
The AND gate 53 outputs an AND signal (read pulse) c of output a of
the comparator 52 and select signal b, and each transistor 42 of
the corresponding row line is turned on. The AND gate 54 outputs an
AND signal of output a of the comparator 52 and signal V.sub.g, and
each transistor 56 is turned on. Consequently, a current is
supplied to the bolometer film (detecting portion 41) from the
current source 55 by way of the transistor 42. As a result, a
voltage is produced at one terminal of the bolometer film, and this
terminal voltage is supplied to the capacitor 58 by way of the
transistors 42 and 56. At this time, depending on the incident
amount of infrared ray in each detecting portion 41, the terminal
voltage of the bolometer film varies. The temperature of the
bolometer film gradually rises by self-heating. In this example,
since the TCR of the bolometer film is negative, the terminal
voltage of the bolometer film decreases gradually with increasing
temperature. Accordingly, the output voltage V.sub.ob of the
transistor 57 corresponding to the insensitive pixel column line
decreases along with the lapse of time.
[0085] When the voltage V.sub.ob becomes equal to the reference
voltage V.sub.c, the output a of the comparator 52 changes from
high level to low level. Therefore, the outputs of the AND gates 53
and 54 also change from high level to low level. As a result, the
transistors 42 and 56 are turned off, and signal reading from the
detecting portion 41 is terminated. Thus, each capacitor 58 is
charged with a voltage corresponding to the voltage signal from
each detecting portion 41, that is, the voltage corresponding to
the incident amount of infrared ray to each detecting portion
41.
[0086] After the select signal b and V.sub.g became low level,
reading of the voltage charged in the capacitor 58 is started.
First, the column select circuit 60 supplies a select signal e to
the corresponding transistor 59, and the charged voltage of the
corresponding capacitor 58 is read out through the selected
transistor 59. In succession, the column select circuit 60 supplies
a select signal f to the corresponding transistor 59, and the
charged voltage of the corresponding capacitor 58 is read out
through the selected transistor 59. Similarly, thereafter, each
capacitor voltage for one row line is read out sequentially.
[0087] When reading of capacitor voltage for one row line is over,
a reset signal V.sub.res is applied again in each transistor 57,
and the signals are detected and read out similarly in the next
line.
[0088] In this embodiment, by setting the reference voltage V.sub.c
of the comparator 52 for each device, the read pulse width is
optimized. Therefore, if the peak temperature of TCR fluctuates
between devices or between lines due to variations of
characteristic of bolometer film, the infrared ray can be detected
near the peak temperature. Hence, even if using a bolometer film
material narrow in a temperature range where a large value of TCR
is obtained, the infrared ray can be detected securely at high
precision.
[0089] Specific examples of this embodiment will be explained
below.
EXAMPLE 1
[0090] A thin film of Sm.sub.1-xA.sub.xNi.sub.yO.sub.3-d was
fabricated by a molecular beam epitaxy (MBE) method.
[0091] FIG. 10 schematically shows a configuration of a molecular
beam epitaxy apparatus.
[0092] As shown in FIG. 10, gas in a vacuum chamber 81 is exhausted
by a cryopump. A substrate holder 82 is provided in the vacuum
chamber 81, and a substrate 83 is placed on the substrate holder
82. The substrate holder 82 is heated by a heater 84.
[0093] Opposite to the substrate 83, plural Knudsen cells 85 are
disposed, and a cell shutter 86 is provided with each Knudsen cell
85. Each Knudsen cell 85 is filled with constituent element of thin
film formed in the following examples, that is, La, Al, Sm, Ni, Bi
and Nd. Further, to obtain a thin oxide film, pure ozone gas
(O.sub.3 gas) stored in a liquid ozone bath 87 is injected from a
nozzle 88, and supplied to the substrate 83. To form a proper thin
film of Sm.sub.1-xA.sub.xNi.sub.yO.sub.3-d, Ni.sup.3+ is needed,
and a strong oxidizing condition is required. In this example,
Ni.sup.3+ could be successfully produced by using pure ozone gas
which has very strong oxidizing power. The substrate temperature
was 500 to 750.degree. C. in the film forming process in this
example. In the process of cooling to 200.degree. C. after forming
the film, ozone gas was supplied continuously to oxidize
sufficiently.
[0094] First of all, the film forming condition for obtaining a
single phase film of SmNi.sub.yO.sub.3-d was studied. As a result
of X-ray diffraction, at substrate temperature of 500.degree. C.,
desired crystal structure was not obtained, and only an amorphous
structure was produced. In the case of epitaxial growth at
substrate temperature of 550 to 750.degree. C., a single phase film
of SmNi.sub.yO.sub.3-d was formed. When a single crystal substrate
of LaAlO.sub.3 (100) was used in this substrate temperature range,
metal-insulator transition occurred, and a large value of
.vertline.TCR.vertline. was obtained.
[0095] FIG. 11 shows the temperature dependence of resistivity (a)
and temperature dependence of TCR (b) in the case of typical
material. In this material, the metal-insulator transition was
obtained at about 410 K. The maximum absolute value of TCR exceeds
6%/K. This value is two times or more of the TCR value of the
conventional vanadium oxide.
[0096] FIG. 12 shows the temperature dependence of TCR in the case
of using LaAlO.sub.3 substrate. At substrate temperature of 550 to
750.degree. C., .vertline.TCR.vertline. exceeded 3%/K.
[0097] The dependence of TCR on the ozone gas flux was studied.
FIG. 13 shows the dependence of TCR on the O.sub.3 molecular
flux/Ni flux ratio in the case of using LaAlO.sub.3 substrate. When
the O.sub.3 molecular flux was 30 times or more of the Ni flux,
.vertline.TCR.vertline. exceeded 3%/K. At this time, the O.sub.3
molecular flux on the substrate was 1.7 to 2.2.times.10.sup.-5
mol.sec.sup.-1.m.sup.-2. The method of oxidization includes, beside
the ozone gas oxidizing method, a method of generating oxygen
plasma by high frequency discharge or electron cyclotron resonance,
and oxidizing by using this oxygen plasma. In this case, when the
active oxygen flux is 30 times or more of the Ni flux, a high value
of .vertline.TCR.vertline. can be obtained.
[0098] When SrTiO.sub.3 or NdGaO.sub.3 was used as the substrate,
metal-insulator transition was not obtained, and the value of
.vertline.TCR.vertline. was small. By contrast, in the case of
depositing LaAlO.sub.3 film on the substrate as underlying layer
about 100 nm by MBE method, a high value of .vertline.TCR.vertline.
was obtained same as in the case of LaAlO.sub.3 single crystal
substrate.
[0099] In a compound having perovskite structure, the composition
ratio of site A element and side B element is usually 1, but when a
thin film is fabricated, the composition ratio is often deviated
from 1. FIG. 14 shows the Ni/Sm composition ratio dependence of
TCR. When the Ni/Sm composition ratio is 0.9 or less, the value of
.vertline.TCR.vertline. drops. To obtain a value of
.vertline.TCR.vertline. of 3%/K or more, 0.9<y is needed in
Sm.sub.1-xA.sub.xNi.sub.yO.sub.3-d.
[0100] By replacing part of Sm with Bi, effects on T.sub.MI were
studied. As a result of experiment, supposing Bi substitution
amount to be x in Sm.sub.1-xBi.sub.xNi.sub.yO.sub.3-d, T.sub.MI was
found to be approximated by the following formula.
T.sub.MI(K)=-1170x+403
[0101] To be applicable to a non-cooled type sensor,
T.sub.MI>300 K should be required. For this purpose, x must be
less than 0.09. Assuming to hold the device at temperature
T.sub.S=300 K, and considering self-heating of bolometer film, 320
K.ltoreq.T.sub.MI<350 K is preferred. Hence, 0.045<x<0.071
is desired.
[0102] In this example, Knudsen cells are used as molecular beam
supply source, but an electron beam evaporation method may be also
used as means of molecular beam source. A thin film can be also
formed by a method of supplying molecular beam of organic metal
from Knudsen cell or gas source nozzle. In the example, the thin
film was formed by molecular beam epitaxy method, but it may be
also formed by sputtering method, laser ablation method, or
chemical vapor deposition method (CVD method). In particular, the
organic metal CVD method is preferable because it is suited to mass
production.
EXAMPLE 2
[0103] A thin film of Ca.sub.2RuO.sub.4 was formed by RF sputtering
method.
[0104] Using a Ca.sub.2RuO.sub.4 sinter target of 4 inches in
diameter, RF power of 60 W was applied. The sputtering gas was a
mixed gas of Ar 90%+O.sub.2 10%, the flow rate was 33 sccm, and the
pressure was 1 Pa. The substrate temperature was room
temperature.
[0105] Substrates were SrTiO.sub.3 (100) single crystal substrate,
NdGaO.sub.3 (001) single crystal substrate, and LaAlO.sub.3 (100)
single crystal substrate. As a result, only in the case of using
LaAlO.sub.3 substrate and annealing at temperature of 975.degree.
C. or more after forming the film, Ca.sub.2RuO.sub.4 having a
desired K.sub.2NiF.sub.4 type crystal structure was obtained.
[0106] FIG. 15 shows the heat treatment temperature dependence of
Ca.sub.2RuO.sub.4 (002) peak intensity of X-ray diffraction. When
heated at 990.degree. C. or more and 1050.degree. C. or less,
Ca.sub.2RuO.sub.4 of excellent crystallinity was obtained. If
heated at 1050.degree. C. or more, Ca.sub.3Ru.sub.2O.sub.7 was
mixed as impurity phase. When heated at less than 990.degree. C.,
the amount of CaRuO.sub.3 impurity phase increased. When heated at
990.degree. C. or more and 1050.degree. C. or less, metal-insulator
transition was obtained. FIG. 15 shows results of heat treatment in
a mixed gas of 99.5% of nitrogen gas and 0.5% of oxygen gas.
[0107] FIG. 16 shows the oxygen partial pressure dependence of
Ca.sub.2RuO.sub.4 (002) peak intensity of X-ray diffraction. When
the oxygen concentration was 0.05 to 1%, Ca.sub.2RuO.sub.4 with
excellent crystallinity was obtained. As a result of measurement of
temperature dependence of resistance, metal-insulator transition
was obtained at the oxygen concentration of 0.05% and 0.5%, but
metal-insulator transition was not obtained at oxygen concentration
of 1%. Therefore, a proper oxygen concentration should be 0.05% or
more and less than 1%. FIG. 15 shows results of heat treatment at
temperature of 1010.degree. C.
[0108] FIG. 17 shows the temperature dependence of resistivity (a)
and temperature dependence of TCR (b) in the case of a typical
material. In this material, the metal-insulator transition was
obtained at about 248 K. In this material, the heat treatment for
obtaining a desired crystal structure was conducted in 0.5% oxygen
atmosphere for 30 minutes at 1010.degree. C. As a result of
chemical analysis of this material, the atomic ratio of Ca/Ru was
1.392. Owing to lack of Ca, the T.sub.MI was lower.
[0109] To lower the T.sub.MI, it was found for the first time that
it is effective to lose part of Ca. In Ca.sub.2-xRuO.sub.4-d,
supposing the loss amount to be x, T.sub.MI was found to be
approximated by the following formula.
T.sub.MI(K)=-179x+357
[0110] To be applicable to a non-cooled type sensor,
T.sub.MI>300 K should be required. For this purpose, the loss
amount x must be less than 0.32. Assuming to hold the device at
temperature T.sub.S=300 K, and considering self-heating of
bolometer film, 320 K.ltoreq.T.sub.MI.ltoreq.- 350 K is preferred.
Hence, 0.04.ltoreq.x.ltoreq.0.21 is desired.
[0111] To enhance the characteristic by decreasing the amount of
CaRuO.sub.3 impurity phase, what is important is the process
condition and sample state before the heat treatment for obtaining
a desired crystal structure. To achieve metal-insulator transition
by obtaining Ca.sub.2RuO.sub.4 with excellent crystallinity, the
thin film right after sputtering is preferred to be amorphous.
Accordingly, at the time of sputtering, it is preferred to hold the
substrate at room temperature without heating. If sputtering is
performed with heating the substrate, finally, the amount of
CaRuO.sub.3 impurity phase increases, and the characteristic
becomes worse. By low temperature annealing after sputtering, the
amount of CaRuO.sub.3 of impurity phase is decreased. As a result,
the crystallinity of Ca.sub.2RuO.sub.4 is improved, and a clear
metal-insulator transition can be obtained. This low temperature
annealing is preferred to be conducted at 700 to 800.degree. C.,
for more than 10 hours in 100% oxygen gas atmosphere. If the
annealing time is as short as 3 hours, effects are hardly obtained,
and sufficient effects are obtained in about 24 hours.
[0112] Additional advantages and modifications will readily occur
to those skilled in the art. Therefore, the invention in its
broader aspects is not limited to the specific details and
representative embodiments shown and described herein. Accordingly,
various modifications may be made without departing from the spirit
or scope of the general inventive concept as defined by the
appended claims and their equivalents.
* * * * *